Large Language Models (LLMs) struggle with summarizing extensive product reviews due to limitations in their context windows and the presence of conflicting opinions. To tackle this issue, researchers developed XL-OPSUMM, a scalable framework designed to incrementally summarize reviews. The key feature of XL-OPSUMM is the Aspect Dictionary, which tracks the count of positive, negative, and neutral sentiments for each product feature. This structured approach provides an objective basis for conflict resolution when updating summaries. XL-OPSUMM employs a four-step process for integrating new review data. Initially, reviews are divided into chunks, with the first chunk processed to create a global summary. For subsequent chunks, the framework updates the Aspect Dictionary and generates local summaries. The final step reconciles the local and global summaries using data from the Aspect Dictionary, allowing the system to resolve conflicting opinions effectively. This method enables accurate and comprehensive summaries over time. The framework was tested using the XL-FLIPKART dataset, which comprises around 3,680 reviews per product, significantly larger than the previously used AMASUM dataset. To generate reference summaries for evaluation, the researchers utilized GPT-4-turbo, given its strong performance in similar tasks. XL-OPSUMM outperformed all baseline models, achieving a BooookScore of 85.60, demonstrating its effectiveness in handling large volumes of reviews and summarizing them accurately.